Alzheimer's disease (AD) is projected by 2050 to affect approximately 13.8 million Americans and cost 1.2 trillion dollars per year. Insulin resistance (IR), a reduced cellular responsiveness to insulin, is typically induced by obesity. In late middle-age, IR is associated with deficits in memory and executive function performance and increased AD risk. The applicant, Dr. Auriel Willette, recently found using Magnetic Resonance Imaging (MRI) and Positron Emission Tomography (PET) data that IR in late middle-aged adults is associated with atrophy, lower glucose uptake, and amyloid deposition in prefrontal cortex (PFC), a brain area impacted by early AD that subserves executive function. He also previously found that long-term calorie restriction (CR) in aged rhesus macaques lowers IR, reduces PFC atrophy associated with IR, and improves cognitive performance. The goal of this project is to determine if a previously validated intermittent CR diet that reduces IR in late middle-aged adults, composed of five ad libitum days and two consecutive days of consuming 530 kilocalories (5-2 CR), improves executive function performance and changes PFC functional MRI (fMRI) activation during executive tasks. To accomplish this project, Dr. Willette requires additional supervised research training in human 5-2 CR, clinical trials, task-based fMRI neuroimaging, cognitive testing, AD neuropathology, and endocrinology specific to insulin signaling. Dr. Willette's primary mentor at the National Institute on Aging, Dr. Mark Mattson, is a leader in CR, neuroscience, and conducting CR clinical trials. Dr. Willette will work with a cross-disciplinary team of intramural and extramural co-mentors to gain the required training for performing this study. Specifically, Dr. Willette will add to his two years of clinical experience by being trained in conducting clinical trials through Dr. Josephine Egan, the NIA Clinical Director and Chief of the Laboratory of Clinical Investigation, Dr. Mattson, and Dr. Dimitrios Kapogiannis, a tenure-track Neurologist who has conducted several clinical trials. Dr. Willette will develop expertise in task-based fMRI and cognitive testing from Dr. Kapogiannis, Dr. Arthur Kramer, a leader in the field of cognition and aging with over 30 years of neuroimaging experience, and Dr. Richard O'Brien, Chair of Neurology at Johns Hopkins University. For AD neuropathology, Drs. Mattson and O'Brien are established experts in animal models and older adults respectively. For insulin signaling, Dr. Egan is a senior investigator and pioneer in the field of insulin regulation, incretins, and glucose metabolism. In Aims 1 and 2 (K99 mentored phase), 40 cognitively normal, late middle-aged women with IR on a given visit will engage in a battery of executive function tasks outside of the MRI scanner, perform a fMRI color-word Stroop task that induces PFC brain activity, and perform a fMRI energy-dense food preference task that also activates PFC through cognitive appraisal of food. This testing will occur at baseline and after 8 weeks of either 5-2 CR (n=20) or an ad libitum control diet (n=20). Compliance visits will be conducted every 2 weeks. This project will be conducted through the Laboratory of Clinical Investigation at Harbor Hospital, the NIA's clinical research unit. The NIA Scientific Review Committee has approved this project, and through the Intramural Research Program has adequate resources and dedicated personnel time to be completed in a timely manner. Additional training will include coursework in clinical trials, neurobiology of aging, ethics, management seminars, and workshops for prospective independent investigators including grant writing and tenure-track topics. In Aim 3 (R00 independent phase), Dr. Willette will combine his previous and new training to conduct a similar 5-2 CR intervention in aged patients with Mild Cognitive Impairment (MCI) and IR. In addition to cognition and fMRI, he will acquire and process fluorodeoxyglucose data (FDG-PET) to examine resting glucose uptake in PFC. He will gain FDG-PET training during the K99 phase from Dr. Bradley Christian, senior PET researcher and co-director of PET Physics at the University of Wisconsin-Madison. Given that IR is higher in MCI compared to cognitively normal adults, associations with cognition and brain measures may be stronger than in the late middle-aged cohort. The 5-2 CR intervention is expected to improve executive function, stabilize fMRI PFC activity, and increase PFC glucose uptake. These outcomes would have a substantial impact on possible interventions for early AD.